Google Hummingbird is the rewrite of Google’s core search algorithm announced in September 2013. Instead of matching the individual words in a search, Hummingbird was designed to understand the meaning of the whole query, which made Google far better at conversational, long and voice searches. Google said it affected around 90% of searches worldwide.
On this page
What the Hummingbird Google algorithm is
Google announced Hummingbird on 26 September 2013, at an event marking its 15th anniversary, and said it had already been running for about a month. Unlike Panda or Penguin, which were filters added to the existing algorithm, Hummingbird replaced the core engine itself. Google chose the name because the new algorithm was meant to be “precise and fast”.
What Hummingbird changed
- Whole-query understanding. Google began interpreting what a sentence means instead of looking for pages that contain each word.
- Conversational search. Questions phrased the way people speak, including voice searches, started returning direct, relevant results.
- Use of the Knowledge Graph. Hummingbird drew on Google’s database of people, places and things to connect a query to real-world entities.
- Synonyms and context. A page could rank for a search without containing the exact phrase, if it answered the intent.
An example
Take the search “where is the closest place to buy a phone charger”. A keyword-matching engine looks for pages containing “buy” and “phone charger”. Hummingbird understands that “place” means a physical shop, that “closest” depends on where the searcher is, and returns nearby stores.
Hummingbird compared with other Google updates
| Update | Year | Focus |
|---|---|---|
| Panda | 2011 | Thin and low-quality content |
| Penguin | 2012 | Manipulative links |
| Hummingbird | 2013 | Understanding the meaning of queries |
| Pigeon | 2014 | Local search results |
| RankBrain | 2015 | Machine learning for unfamiliar queries |
| BERT | 2019 | Language understanding in context |
What Hummingbird means for SEO today
Hummingbird has been built on many times, but the principle it introduced still shapes how Google and AI search engines work. In practice:
- Write for topics and intent, not single keywords.
- Answer the questions people actually ask, in clear language, near the top of the page.
- Cover a subject fully, including related terms and follow-up questions.
- Use structured data and consistent naming so Google can connect your content to entities. See entity-based SEO.
- Stop producing separate pages for every keyword variation; one strong page usually ranks for all of them.
Was Hummingbird a penalty?
No. Hummingbird did not target a type of site and few sites reported sudden losses when it launched. Its effect was gradual: pages that matched keywords without answering the question lost ground to pages that did.
Frequently asked questions
What is the Google Hummingbird algorithm?
Hummingbird is the 2013 rewrite of Google’s core search algorithm. It was built to understand the meaning of a whole query instead of matching individual keywords.
When was Google Hummingbird released?
Google announced it on 26 September 2013 and said it had been live for about a month before the announcement.
Is Hummingbird still used by Google?
Google’s core algorithm has changed a great deal since 2013, with RankBrain, BERT and other systems added, but the meaning-based approach Hummingbird introduced remains the foundation.
How do I optimise for Hummingbird?
Write content that answers real questions in natural language, cover topics in depth and focus on search intent instead of repeating exact-match keywords.
Want experts to handle this for you?
Our specialists turn these fundamentals into rankings, leads and revenue. Start with a free, no-obligation audit of your site.